FET driver

Post a schemao we can really understand your idea :-)

Do you mean "use an op-amp as a comparator"?

Yes and no. If I understand what you're trying to do there are several drawbacks to your scheme:

First, there are a lot of excellent, inexpensive, dedicated gate-driver chips out there. There are parts that don't cost much more than a jelly- bean op-amp, and don't require any external resistors at all. Moreover, there are high-side drivers that let you use a pair of N-channel FETs as a half-bridge, with only an external cap.

The guys designing gate drivers from discrete components are working in markets where margins are very slim, assembly labor is very cheap, and volumes are high enough to justify the extra engineering expense.

Second, op-amps are SLOW compared to gate drivers. Sure, if you don't need fast switching speed you can use a general-purpose device -- but if you DO need fast switching speed, then you're hardly going to be buying jelly-bean op-amps.

Third, it's generally not a good idea to use op-amps as comparators. Some work OK, some don't do well when there's a large differential voltage across their inputs. If you want a comparator, use a comparator.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

An opamp can make a nice on-off mosfet gate driver.

If you run it open-loop, as a comparator, you can get nearly or totally rail-to-rail output swing. Slew rates will be mediocre for slow amps, which generally won't matter and may help.

A series gate resistor is a good idea, 50 ohms sort of range. I've seen mosfets do RF oscillations when driven by an opamp.

Most opamps current limit in the 15 mA ballpark, but some are much higher. That's shouldn't affect the average gate drive situation.

There are nice on-purpose cheap gate driver chips that can deliver amps of gate drive in nanoseconds, which no opamp can do.

Do you mean adding transistors after an opamp?

If you run the opamp open loop, it needs no feedback resistors, just a shared DC voltage something near mid-swing of your logic levels. Two resistors will do that for any number of channels.

Don't use an LM324.

--

Sometimes I wonder how many LM324s are manufactured each year. 


----Android NewsGroup Reader---- 
http://usenet.sinaapp.com/

I just designed-in my first LM324 ever. It's in a spark detector for cotton mills. Cotton mills generally transport cotton from bales through the gin and carding operations via air ducts. You can transport

2000 kg of cotton fibres per hour (in small clumps) through a 30 cm diameter duct. It has to be done at low humidity, or the cotton will stick to everything, so there's lots of static electricity. Dry cotton is very flammable, which makes fire prevention a priority.

Lots of cotton is processed in poor countries, so the BOM cost has to be rock-bottom: hence the 324. It's the temperature sensor, differentiator, buffer amp, and output comparator. (There's a TLV271I for the TIA and bias generator.)

Major discovery: Hamamatsu (yes, Hamamat$$u), makes a $5 InGaAs photodiode. That's triple the cost of the rest of the circuitry, even at 1k-piece distributor prices. Because it's a safety device, I'm spending almost as much money on the protection networks as on the sensor itself.

Designing things for super low cost and high performance is actually pretty fun.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

On a sunny day (Thu, 3 Dec 2015 13:06:43 -0500) it happened Phil Hobbs wrote in :

Cannot you use radio to detec tsparks? I build a lightning detector, basically a ferrite core tuned to some MHz, uses a PIC 12F comparator and processing: circuit diagram:

whole project, hex file etc

Old project never finished, any old long wave radio is better... but it worked, you do not need the crystal, some PICs have internal osc. Could work with LM324 too.

Maybe do not tune, RF choke perhaps.

These aren't electrical sparks, they're burning clumps of cotton.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

The original National LM324 was terrible as a comparator. If one opamp slewed to a rail, the other outputs could glitch seriously. They shared some current sources or something. I think that other manufacturers may have fixed that in their clones.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

They shared the power rails for one, could be other issues..

Recent stock I have played with shows that issue no longer noticeable. I have noticed that the outputs do seem very sensitive to minor very abuse, they tend to short very easy. I never noticed that issue so bad with old stock, they would take a little abuse before giving up.

The new stock would be NOPB of course. Could it be they are using a different process and maybe a slightly different design ?

I guess Jim T would know better.

Jamie

Well, maybe that's your problem. You are just playing.

What?

Perhaps, if he could make sense of your word salad. Are you in the world of the living?

John S,

You are being a jerk.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

that's often the way.

NT

Some long time ago, I designed some flame detectors, for shutting off the fuel oil flow in shipboard boilers, if the flame went out. Spraying a bunch of unburned oil into a hot boiler can have seriously bad consequences if anybody tries to relight it.

The detector was a UV-sensitive, solar blind, quartz bulb, gas-filled discharge tube that looked like a small light bulb. In full room light, it would fire from a match flame six feet away. I think Hamamatsu made some of them.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Phil, can I ask how many detector's per 100' (or not, NDA) and what happens after you detect a cotton fireball? (sounds fun.) George H.

You hit a bloody great solenoid that runs a diverter door downstream, to dump the burning bits to the great outdoors. Some systems have big dry-chemical extinguishers as well. CO2 pressure does a lot of the work for you--it's amazing the amount of powder you can squirt through a

3-inch valve with a liquid CO2 tank behind it.

Apparently some folks who have opened an access panel without disabling the automatics have wound up looking like the Abominable Snoman.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

This is most likely a crazy idea, but I was thinking we could have some code of conduct here. I would love it, if we would all behave like ladies and gentlemen. Sure sometime you need to tell other's that you think they are mistaken. There is no reason to be rude about it, and if they don't listen so what? (One nice thing about electronics/ instruments is the success or failure is there at the output.)

Since you and JT have stopped snipping, things have been nothing but better. For me, I see SED as this huge resource that I can tap into. And tomorrow I'll have a problem or question about X. I don't know what it will be or who here might have an answer. Isn't that enough of a motivation to be nice? It costs next to nothing, it's like using your turn signals when you drive. (I always use my turn signals, it's polite.)

The vast majority of people here seem totally decent. (sorry for the rant)

George H.

We should hire Guy Macon back to oversee it. ;)

The main thing for everybody to remember is that the guy who gets the last word doesn't win. Assuming anybody besides the Rockem-Sockem Robots is still following the exchange, which is doubtful, the last one standing is usually just slowest on the uptake.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

The pattern with both JF and JT is that I say something objective (about things, not about them) and they react by going personal. Look it up.

I recently made a small improvement to JF's 3-phase sine generator, and he then accused me of all sorts of personality disorders. Makes no sense.

That's what John S just did, forked a technical discussion into insults.

JT got mad at me when I pointed out that you can buy opamps with 1000 times the slew rate of his "fast slew" amp. It's true.

One pattern to note: the technically competent people here don't start or join the flame wars. And vice versa.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

The other things that 324s usually do (aside from the ghastly crossover distortion) is go bezerk if any one of the inputs gets pulled a bit below ground.